Should developers of a new product that requires collaboration focus on internal integration of departments using cross-functional teams or on external linkages with suppliers and customers? While both types of integration can help, our recent international study of new product development (NPD) practices suggests that internal integration will affect time and product performance differently than external integration.
Our study analyzed survey responses to questions about the procedures and performance measures of the NPD programs at 250 small- to medium-size Machinery Manufacturing and Electrical Equipment firms (codes SIC 35 and SIC 36) in four countries: Italy, Germany, Japan, and the United States. Ninety percent of the firms employed fewer than 3,000 people. All firms were randomly selected but the population was chosen because of the propensity to perform NPD and its importance to strategic business success in the industry. Individual respondents included CEOs and senior managers at the VP or programlevel of the NPD, R&D or design department.
The study was conducted at the program level whereas much NPD research focuses at the project level. Programbased studies can increase generalizability by determining the average effects of total program efforts on firm success. We defined the " development program " as the set of new products developed and launched within the previous three years. All the companies had at least five projects as part of their overall NPD program. This provided useful evidence about the long-term relationships among variables of interest.
Significant Differences
The study's overall findings suggest that the use of collaborative NPD practices such as customer involvement, supplier involvement and cross-functional integration are found across the world although we did find some signifi- cant differences in focus. Specifically, American firms were more likely to use internal integration practices and this translated into better performance. Also, the use of internal integration is still emphasized more than external integration, which is newer to the management of NPD ( 1 - 3 ). Regression analyses of the data have shown that regardless of country origin these practices have specific, material effects on performance for all the firms surveyed ( 4 ).
Analyses also demonstrate that focusing on specific types of integration may be necessary to achieve specific goals. The diagram (next page) shows the relationships found among the important elements of NPD collaborative strategy. For example, internal integration, which consists of traditional cross-functional management practices, directly influences a firm's innovation capabilities. The link between internal integration practices and performance as measured by both time and product success depends on the creation of these innovation capabilities.